4778
J. Org. Chem. 1991,56,4778-4783
termined by the equation ks/kH = (PS/PH)(HO/S“‘). Here, Ps/PH 702 cm-l; MS m / z 265 (M+ - C2Ha,54), 205 (361,149 (1001, 145 (811,115 (32),91(27),77 (34),59 (40).Anal. Calcd for C I ~ H ~ O ~ :is the molar ratio of the reaction products formed from the competing styrenes and H 0 / P is the molar ratio between unC, 65.29; H, 7.53. Found C, 65.03; H, 7.64. S-Ethyl-J-(methsubstituted and substituted styrene before the reaction. In no oxycarbonyl)-5-phenyl-y-butyrolactone [10 (R’= OMe, R” cases the difference between the values determined by the two = Et, Ar = C6H6)]was also recovered in 59% of yield (1.3:l methods exceeded 3%. mixture of two diastereoisomers): ‘H NMR (200 MHz) referred to the main stereoisomer 6 7.3 (tight m, 5 H), 3.78 (8, 3 H), Acknowledgment. We are very grateful to Prof. B. 3.54-3.43 (four peaks, 1 H), 2.92-2.63 (m, 2 H), 2.08 (9,J = 7.1 Hz, 2 H), 0.83 (t,J = 7.1 Hz, 3 H); referred to the other stereoGiese for making available a manuscript to us before isomer 6 7.35 (tight m, 5 H), 3.67 (a, 3 H), 3.84-3.73 (four peaks, publication. We also thank the Miniitero d e b Universita’ 1 H), 2.97-2.62 (m, 2 H), 1.98 (9,J = 7.1 Hz, 2 H), 0.82 (t, J = e della Ricerca Scientifica e Tecnologica and the Consiglio 7.1 Hz, 3 H); IR (film) 3095-3005,2975-2850,1780,1740,1602, Nazionale delle ricerche (C.N.R.) for financial support. 1450, 1195,703 an-’. Anal. Calcd for c14Hleo4: C, 67.73; H, 6.49. Registry No. 5b, 2039-93-2; 5c, 17498-71-4;5d, 5676-29-9; 8 Found C, 67.97; H, 6.55. (R = R’ = Me, R” = H, Ar = (&Ha), 13463-61-1;8 (R= R’ = Me, Kinetic Studies. The relative rate constants of Z- and RR” = H, Ar = p-MeC6H4),134391-03-0; 8 (R = R’ = Me, R” = substituted styrenes with respect to styrene in the CAN-promoted H, Ar = p-ClC&), 134391-04-1; 8 (R = R’ = Me, R” = H, Ar oxidative addition of 2,4-pentanedione, methyl 3-oxobutanonate, = m-ClC&$, 134391-05-2; 8 (R = R’ = Me, R” = H, Ar = mand dimethyl malonatz were determined by the competitive N02C6H4),134391-06-3; 8 (R = R’ = R” = Me, Ar = C&), method. Kinetic data for the oxidative addition of dimethyl 54023-32-4; 8 (R = R’ = Me, R” = Et, Ar = C&), 134391-07-4; malonate to Zsubstituted styrenes were available from a previous 8 (R = R’ = Me, R” - i-Pr, Ar = C&), 134391-08-5; 8 (R = R’ work.I0 To a solution of CAN (1-3 mmol) in 4 mL of solvent, a = Me, R” - tBu, Ar = c&), 134391-09-6;8 (R = Me, R’ = OMe, solution of dicarbonyl compound (ca. 0.5 mmol), substituted R” = H, Ar = C&), 134391-10-9; 8 (R = Me, R‘ = OMe, R” = styrene (0.6-1.0 mmol), styrene (0.6-1.0 mmol), and m-chloroH, Ar = m-NO2C6H4),134391-11-0;9 (R = R’ = OMe, R” = i-Pr, toluene (ca0.75 “01) as an internal standard in the same solvent Ar = C a s ) , 134391-12-1;9 (R = R’ = OMe, R” = tBu, Ar = C&, (1.0 mL) was added at 20 “C. The mixture was allowed to react 134391-16-5;9 (R= R’ = OMe, R” = Me, Ar = C&), 134391-18-7; 2-5 min (reactions with 2,4-pentanedione and methyl 3-oxo9 (R = R’ = OMe, R” = Et, Ar = C&), 134391-21-2; 10 (R’= butanoate) or 40 min (reactions with dimethyl malonate). The OMe, R” = i-Pr, Ar = C&,) (isomer l), 134391-13-2; 10 (R‘= mixture was poured into water (50 mL) and extracted with hexane OMe, R” = i-Pr, Ar = C6H6)(isomer 2), 134391-14-3; 10 (R’= (3 X 25 mL). The coUacted organic phases were washed with water OMe, R” = t-Bu, Ar = C&), 134391-17-6; 10 (R’ = OMe, R” = (75 mL), dried with anhydrous sodium sulfate, and analyzed by Me, Ar = C a d (isomer l),134391-19-8;IO (R’ = OMe, R” = Me, GLC after suitable dilution. The relative rate constants were Ar = C6H6)(isomer 2), 134391-20-1; 10 (R’ = OMe, R” = Et, Ar determined by the equation ks/kH = log (S”’/S)/log (Ho/H) where = C6H5)(isomer l),134391-22-3; 10 (R’ = OMe, R” = Et, Ar = SO/S and Ho/H are the molar ratios of substituted styrene and C&a) (isomer 2), 134391-23-4;CAN, 15078-94-1;a-methylstyrene, styrene, respectively, before and after the reaction. In all cases 98-83-9; ethyl phenyl ketone, 93-55-0; isopropyl phenyl ketone, the amount of reacted styrenes corresponds (5%)to the amount 611-70-1; tert-butyl phenyl ketone, 938-16-9; methyltriphenylof dicarbonyl compound consumed in the reaction. An identical phosphonium iodide, 2065-66-9; lithium diphenylcuprate, procedure was followed for the reactions with dimethyl malonate 23402-69-9; pivaloyl chloride, 3282-30-2; 2,4-pentanedione, 123where CAN was used in a 6:l molar ratio with respect to the 54-6;pmethylstyrene, 622-97-9;styrene, 100-42-5; pchlomtyrene, dicarbonyl compound, as well as for the reactions carried out in 1073-67-2; m-chlorostyrene,2039-85-2; m-nitrostyrene, 586-39-0; the presence of lithium perchlorate. In the reactions with 2,4methyl 3-oxobutanoate, 105-45-3; 5-isopropyl-5-phenyl-ypentanedione, competitive experiments were also carried out using butyrolactone, 134391-15-4. a large calculated excess of styrenes and the relative rates de-
Stereochemical Control of Microbial Reduction. 17. A Method for Controlling the Enantioselectivity of Reductions with Bakers’ Yeast Kaoru Nakamura,* Yasushi Kawai, Nobuyoshi Nakajima,’ and Atsuyoshi Ohno Institute for Chemical Research, Kyoto Uniuersity, Uji,Kyoto 611, Japan Received September 25, 1990 (Revised Manuscript Received April 13, 1991) The stereoselectivity of the bakers’ yeast catalyzed reduction of @-ketoesters to optically active @-hydroxy esters can be controlled by the introduction of a third reagent. To gain ineight into the mechanism of this enzymatic reduction, @-hydroxyester oxidoreductases were isolated from the cells of raw bakers’ yeast. Four dominant competing enzymes were isolated, purified, and characterized. Among these, two reduce &keto esters stereospecifically to the corresponding D-&hydroxy esters. The other two afford the L-hydroxy esters. The rates of enzymatic reduction were determined in the presence and absence of the additives.
Introduction The development of methods for the synthesis of optically active compounds has become one of the most important goals in the fields of organic chemistry and bio-
chemistry. In recent years, there have been dramatic developments in the asymmetric synthesis of organic compounds.’ In particular, biological methods have been extensively developed within the last decadeq2
‘Department of Food and Nutrition, Okayama Prefectural Junior College, Okayama 700, Japnn.
(1)Morrison, J. D., Ed. Asymmetric Syntheaia; Academic P w : New Yo&, 1985; Vola. 2-5.
0022-3263/91/1956-4778$02.50/00 1991 American Chemical Society
J. Org. Chem., Vol. 56, No.15, 1991 4779
Stereochemical Control of Microbial Reduction Scheme I Bakers'Yeasl
Rm
DO
R
*
a
0
R
0
OH
2
Bakers'Yeasi ciJOEIc
0
OR'
R
L
Bakers' yeast reduces various 8-keto esters t o the corresponding optically active &hydroxy esters. These compounds can be utilized as chiral building blocks in the synthesis of natural p r ~ d u c t s . ~However, the stereoselectivity of such reductions is not always high, and a particular microbe does not necessarily afford a product of the desired configuration. A number of researchers have tried to control and improve the stereoselectivity of the reduction by screening microbes? modifying the structure of the substrate: and modifying the reaction conditions! Recently, we reported a quite new and useful method of reduction with bakers' We found that the introduction of a third reagent into the reaction system changes the stereoselectivity of the reduction and allows a product of the desired configuration t o be obtained in high enantiomeric excess. Thus, the introduction of allyl alcohol7 or an c+unsaturated carbonyl compounds shifts the stereoselectivity of the reduction toward the D isomer, whereas the introduction of ethyl chloroa~etate~ favors the formation of the L isomer (Scheme I). T h e method is useful because the stereoselectivity can be easily controlled without screening microbes or modifying the structure of the substrate. Because it would be valuable to elucidate the mechanism by which the stereoselectivity of microbial reduction is controlled by an achiral additive, several &hydroxy ester oxidoreductases were isolated from the cells of raw bakers' yeast and the effects of additives on the activities of the enzymes were studied. The rates of enzymatic reduction were determined in the presence and absence of the additives. T h e degree of stereochemical control was substantiated by t h e results of the rate studies.
Experimental Section Chromatography. Analytical GLC was performed with a 25-m PEG-20M Bonded column at 130 OC. Preparative GLC was performed at 150 "C with a 1.5-m PEG-20M column. HPLC was performed with a 4.6 mm X 150 mm Wakosil 5SIL column. (2) For reviews, see: (a) Sih, C. J.; Chen, C.-S. Angew. Chem., In= Engl. 1984,23,570. (b) Whitesides, G. M.; Wong, C.-H. Angew. Chem., Int. Ed. Engl. 1986,24,617. (c) Simon, H.; Bader, J.; G h t e r , H.; Neuman, S.;Thanon, J. Angew. Chem.,Int. Ed. Engl. 191,24,539. (d) Jonea, J. B. Tetrahedron 1986,42,3351. (e) Toone, E. J.; Simon, E. S.; Bedm k i , M.D.; Whitesides, G. M. Tetrahedron 1989,46,5365. (0 Wong, C.-H. Science ISSS,244,1145. (g) Klibanov, A. M. Acc. Chem. Res. 1990, 23, 114. S e d , s. Synthed8 1990, 1. (3) (a) Nakamura, K.; Kitayama, T.; Inoue, Y.; Ohno, A. BulZ. Chem. SOC.Jpn. ISSO, 63,91-96. (b) For a review, see: Mori, K. Tetrahedron l S S S , 4 , 3233-3298. (4) (a) Dauphin, G.; Fauve, A,; Veschambre, H. J . Org. Chem. 1989, 64, 2238-2242. (b) Fauve, A.; Veschambre, H. J. Org. Chem. 1988,53, 5216-5219. (c) Nakamura, K.; Miyai, T.; Fukushima, K.; Kawai, Y.; R.ameah, B. B.; Ohno, A. Bull. Chem. SOC.Jpn. 1990, 63, 1713-1715. (5) (a) Zhou, B. N.; Gopalan, A. S.; VanMiddlesworth, F.; Shieh, W. R.;Sih, C. J. J. Am. Chem. SOC.1989,106, 5926-5926. (b) Nakamura, K.; Miyai, T.; Nagar, A.; Oka,S.; Ohno, A. Bull. Chem. SOC.Jpn. 1989 62,11791187. (c) Nakamura, K.; Miyai, T.; Ushio, K.; Oka, S.; Ohno, A. Bull. Chem. SOC.Jpn. 1988,61,2089-2093. (d) Nakamura, K.; Miyai, T.; Nozaki, K.; Ushio, K.; Oka, S.; Ohno, A. Tetrahedron Lett. 1986,27, 3155-3156. (6) (a) Fauvs, A,; Veschambre, H.; Tetrahedron Lett. 1987, 28, 5037-6040. (b) Nakamura, K.; Higaki, M.; Ushio, K.; Oka, S.;Ohno, A. Tetrahedron Lett. 1986,26, 4213-4216. (c) Nakamura, K.; Inoue, K.; Ushio, K.; Oka, 5.;Ohno, A. J . Org. Chem. 1988,63, 2589-2593. (7) Nakamura, K.; Inoue, K.; Ushio, K.; Oka, S.;Ohno,A. Chem. Lett. 1987.679-882. - - - ., - . - - - -. (8) Nakamura, K.; Kawai, Y.; Oka, S.;Ohno, A. Bull. Chem. SOC.Jpn. 1989,62,876-679. (9) Nakamura, K.; Kawai, Y.;Ohno, A. Tetrahedron Lett. 1930,31, 267-270.
Materials. Ethyl 4-chloro-3-oxobutanoatewas purchased from Tokyo Kasei Kogyo Co. (R)-(+)-a-Methoxy-a-(trifluoromethy1)phenylacetic acid (MTPA) was purchased from Aldrich Chemical Co. Bakers' yeast and a MW-Marker (a mixture of cytochrome c, 12.4 kDa; adenylate kinase,32 kDa; enolase, 67 ma; lactate dehydrogenase, 142 kDa; glutamate dehydrogenase, 290 kDa) were purchased from Oriental Yeast Co. DEAE-Toyopearl 650 and Butyl-Toyopearl 650 were purchased from Toeoh Co. The Superose 12 column was obtained from Pharmacia Fine Chemicala The Ultra Pack PEI (polyethylene imine), CE (carboxy ethyl), and HIC (HI-propyl) columna were purchased from Yamazen Co. Cellurofie GLC 2000 was purchased from Seikagaku Kogyo Co. Asahipack GS-510 was obtained from Asahi Chemical Industries Co. Centricell ultrafiiters (10000 NMWL) were purchased from Polysciences, Inc. Other reagents were purchased from Nacalai Tesque Co. The basic buffer consisted of 10 mM potassium phosphate (pH 7.00), 0.05% 2-mercaptoethanol, 1 mM dithiothreitol, and 1mM (phenylmethy1)sulfonyl fluoride. Enzyme Assay. A 50-pL aliquot of chromatographic fraction (vide infra) was added to 3.00 mL of a solution of 0.10 M potassium phosphate buffer (pH 7.0) that also contained ethyl 4-chloro-3oxobutanoate (0.98 mM) and NADPH (0.09 mM). The rate of reaction was determined spectrophotometrically at 30 OC by following the decrease in absorbance of NADPH at 340 nm. One unit of enzyme activity was defined as the amount of enzyme that catalyzed the oxidation of 1 pmol of NADPH per minute at 30 OC under the these conditions. Determination of Optical Purity of the Product. To a stirred solution of the product 8-hydroxy ester (0.03 mmol) and benzene (1 mL) were added, in order, pyridine (1.5 mmol) and (R)-(+)-a-methoxy-a-(trifluoromethyl)phenylacetylchloride (0.2 "01). The mixture was stirred Overnight. EtOAc and H20 was then added. The two liquid layers were separated. The organic layer was washed (1M aqueous HC1, saturated aqueous NazC08, and brine), dried (NazSO4),and concentrated to give the corresponding (R)-MTPAester in more than 95% yield in all instancee. The enantiomeric excess (ee) of the hydroxy ester was determined by HPLC (hexane-EtOAc, 301) analysis of the corresponding (R)-MTPA ester. Purification of the &Hydroxy Ester Oxidoreductases. The enzymes were isolated, purified, and stored at 4 OC. For prolonged storage, the enzymes were frozen in 10% aqueous glycerol. Thus, raw bakers' yeast (1kg) was suspended in 10 mM phosphate buffer (2 L). The suspended cells were disrupted with a Dyno-Mill operating at a flow rate of 30 ml/min. The disrupted mixture was centrifuged at lOOOOOg for 30 min. The supernatant liquid (1.4 L) was dialyzed overnight against 10 mM phosphate buffer (10 L). The buffer was renewed, and dialysis was continued overnight. The dialyzate was concentrated by dialysis against poly(ethy1ene glycol) (MW 20000) overnight. The concentrated dialyzate (500 mL) was then centrifuged at lOOOOOg for 30 min to obtain 400 mL of supernatant liquid. To eliminate nucleic acids and noneffuse proteins, the cell-free solution (400 mL) was applied to a 6 cm X 30 cm column packed with DEAE-Toyopearl650equilibrated with the basic buffer. The proteins were eluted with the basic buffer, which also contained 0-0.4 M linear gradient of KC1 (4 L). The active fractions were collected and concentrated to 250 mL by dialysis against poly(ethylene glycol) (MW 20000). The concentrated enzyme solution was then dialyzed for 1 day against 3 L of the basic buffer with frequent renewal of the buffer. The dialyzate was applied to a 6 cm X 30 cm column packed with DEAE-Toyopearl650equilibrated with the basic buffer. The column was eluted with the basic buffer (2 L) at a flow rate of 2.0 mL/min. A nonadaorbed enzyme (L-enzyme-1) was eluted. Then, elution was continued with a 0-0.3 M linear gradient concentration of KC1 diasolved in the basic buffer (5 L). Fractions of 20 mL were collected. Three enzymes were eluted: at 0.04 M and 0.15 M enzyme-2) KCl. (D-enzyme-l),0.06 M (~-enzyme-2), These active fractions were each concentrated to 80 mL by dialysis against poly(ethy1ene glycol) (MW 20 OOO). Each of the four enzyme solutions was then applied to a 3 cm x 40 cm column packed with Butyl-Toyopearl 650 equilibrated with the basic buffer, which ale0 contained 20% (NH4)fiOo T h e column was eluted with the basic buffer, which also contained
4780 J. Org. Chem., Vol. 56, No. 15, 1991
20% (NH4)2S04,and then with a 2 0 4 % linear gradient concentration of (NH,)#O4 dissolved in the basic buffer (400 mL). Fractions of 9.0 mL were collected. Thoee that exerted activity were combined and concentrated to 0.5 mL by ultrafiitration with a Centricell centrifugal ultrafilter (loo00 NMWL). Then, the D-enzymes were directly subjected to gel filtration column chromatographyto determine their molecular weights and, at the same time, to purify them for rate studies. The solution of L-enzyme-1 or ~-enzyme-2was applied to an Ultra Pack PEI column equilibrated with the basic buffer. The column was eluted with a 0 4 . 5 M linear gradient concentration of NaCl dissolved in the basic buffer. Fractions that exerted activity were combined and concentrated to 3 mL by ultrafiltration. The concentration solution of L-enzyme-1 was applied to an Ultra Pack CE column equilibrated with the basic buffer (pH 6.0). The column was eluted with a 0 4 . 5 M linear gradient concentration of NaCl dissolved in the same buffer. Fractions that exerted activity were combined and concentrated to 1 mL by ultrafiltration. The concentration solution of ~-enzyme-2was applied to an Ultra Pack HIC column equilibrated with the basic buffer, which also contained 20% (NHJ2SOI. The column was eluted with a 2 0 4 % linear gradient concentration of (NH4)&304dissolved in the basic buffer. Fractions that exerted activity were combined and concentrated to 1 mL by ultrafiltration. The solution of L-enzyme-1 or ~-enzyme-2so obtained was applied to a gel filtration column (Cellurofine GLC 2000 equilibrated with the basic buffer which also contained 0.1 M KCl). The column was eluted with the same buffer. Fractions that exerted activity were combined and concentrated to 0.1 mL by ultrafiltration. Determination of the Molecular Weights of the Enzymes. A solution of D-enzyme-1, ~-enzyme-2,or the MW-Marker wae applied to a gel filtration column (Superose 12 equilibrated with the basic buffer, which also contained 0.1 M KCl). After elution with the same buffer (flow rate 0.5 mL/min), the activity of each fraction was assayed as described previously. The molecular weights of D-enzyme-1 and ~-enzyme-2were found to be 25 and 1600 kDa, respectively. A solution of L-enzyme-1, ~-enzyme-2,or the MW-Marker was applied to a gel filtration column (Asahipak GS-510 equilibrated with the basic buffer, which also contained 0.1 M KCl). After elution with the same buffer (flow rate 0.5 mL/min), the activity of each fraction was aasayed as described previously. The molecular weight of both L-enzyme-l and ~-enzyme-2was found to be 32 kDa. Enzymatic Reduction of Ethyl 4-Chloro-3-oxobutanoate. In a glase reaction vessel were placed a concentrated solution of the enzyme (10 units), NADPH (10 mg), G6PDH (glucose 6phosphate dehydrogenase, 50 units, 0.3 mg), G6P (glucose 6phosphate, 300 mg), ethyl 4-chloro-3-oxobutanoate (82 mg, 0.5 mmol), and 0.1 M phosphate buffer (pH 7.0,30 mL). The reaction vessel was shielded from light. The mixture was magnetically stirred for 24 h at 30 OC. Hytlo Super-Cel and M A Cwere added. The mixture was filtered. The solid that was collectad was washed with EtOAc. The combined washings and filtrate were washed (H20and brine), dried (NafiO,), and concentrated under reduced pressure. The residue was purified by preparative gas chromatography (1.5-m PEG column, 160 O C ) to give ethyl 4-chloro-3hydroxybutanoate. The yields of products were 62,80,45, and 84% from the reductions catalyzed by L-enzyme-1, ~-enzyme-2, Denzyme-1, and enzyme-2, respectively. The optical purity and stereochemistry of the products were determined by the method mentioned previously. The ee of the products were, in all instances, more than 99%.
Results Purification and Properties of @-HydroxyEster Oxidoreductases. The @-hydroxyester oxidoreductases were isolated from a cell-free solution of bakers' yeast by anion exchange column chromatography (DEAE-Toyopearl). A typical chromatogram is shown in Figure 1. Each enzyme was further purified by chromatographic methods, which included hydrophobic chromatography
Nakamura et al.
-
L-enzym-2
= 5
.-
400'
5
300'
.-P
a s
200
ALlivily
,... 8 ..... .... - 0.2 .........
-
.... ...'.... ...' ...' ,... 0-ewrIW-2 .....
100 100-
......... .........'
0.1
I
&
0
50
100
0.0 150
200
250
Fraction No.
Figure 1. Elution chromatogram of &hydroxy ester oxidoreductases derived from bakers' yeast obtained with DEAEToyopearl.
Time (min )
Figure 2. Stability of ~-enzyma2in the absence of and presence of bovine serum albumin (BSA). Table I. Enzymatic Reduction of Ethyl 4-Chloro-3-oxobutanoate total activity config of the enzyme M w (kDa) (U) product D-enZyme-1 25 26 D D-enZyme-2 1600 1442 D L-enzyme-1 32 246 L ~-enzyme-2 32 3108 L
ee (k) >99 >99 >99 >99
and gel filtration chromatography. Each of the enzymes preferentially utilizes NADPH as the coenzyme. The total activities of the enzymes are summarized in Table I. The results demonstrate that D-enZyme-1 contributes little (ca. 0.5% ) to the reduction. It would be reasonable, therefore, to assume that almost all of the reduction by bakers' yeast is catalyzed by enzyme-2, L-enzyme-1, and ~-enzyme-2.Hence, only the rates of reduction catalyzed by the three latter were determined. The stereospecificity of each enzymatic reduction was established by reducing in the presence of a G6P ethyl 4-chloro-3-oxobutanoate, (glucose 6-phosphate)-GGPDH (glucose 6-phosphate dehydrogenase) system in order to regenerate NADPH from NADP+.l0 The enantiomeric excess (ee) of the product was determined by HPLC analysis of the corresponding (R)-MTPA ester. ~-Enzyme-2is too labile to yield the product hydroxy ester under the reaction conditions employed (30 "C),whereas, under the same conditions, the (10) Nakamura, K.; Yoneda, T.;Miyai, T.; Urhio, K.;Oh,S.;Ohno, A. Tetrohedron Lett. 1988,29, 2463-2454.
J. Org. Chem., Vol. 56, No. 15, 1991 4781
Stereochemical Control of Microbial Reduction Table 11. Kinetic Constants of Reductions Catalyzed by the Enzymesa D-enzyme-1 D-enzyme-2 L-enzyme-1 tenzyme-2
100'1 I
80 -
0.59 4.54 0.13 0.15
303b 6.20 4.66
aErrors in KM and kat are estimated to be less than Reference 10.
60 -
&lo%.
40
0
20 0
. 0
D-enzyme-1
0-enzyme-1 0-enzyme.2 L-enzyme-? L-enzyme-2
0
0
20
40
60
Temperature ("C)
Figure 5. Thermal stability of the enzymes.
Temperature("C)
Figure 3. Effect of temperature on the activity of the enzymes.
+ 0
01 4
L-enzyme-1 L-enzyme2
I I
5
6
8
7
9
PH
Figure 4. Effect of pH on the activity of the enzymes.
other enzymes do catalyze the reduction and produce the hydroxy ester. However, as shown in Figure 2, cenzyme-2 can be stabilized by bovine serum albumin (BSA). Thus, in the presence of BSA, ~-enzyme-2is able to afford the product, and, furthermore, the rate of reduction could be determined. The data listed in Table I show that all four enzymes reduce 0-keto esters quantitatively and stereoselectively. The molecular weights of the enzymes, as determined by gel filtration column chromatography, are also listed in Table I. It should be noted that the value of the experimentally determined molecular weight of ~-enzyme-2 is accompanied by an experimental error because the exclusion limit of the gel filtration column is 300 kDa. To further characterize the enzymes, the rates of enzymatic reduction were determined. The results are listed in Table 11. The effects of temperature and pH on the enzymatic activity and the thermal stability of each of the enzymes were also determined. The results are shown in Figures 3-5, respectively. The results yielded fundamental information on how to control the stereoselectivity of the reduction by modifying the reaction conditions. The L-enzymes possess similar properties, e.g., molecular weight and stereospecificity,K M and Iz,. However, they
L-enzyme-1
L-enzyme-2
Figure 6. Polyacrylamide disk gel electrophoreses of L-enzyme1 and ~-enzyme-2.
behaved differently during anion-exchange column chromatography. To differentiate the two enzymes, the homogeneity of each was determined by polyacrylamide disk gel electrophoresis.'l As Figure 6 shows, it is apparent that the two possess quite different electrostaticproperties: the surface of L-enzyme-1 is more positively charged than that of ~-enzyme-2. Effect of Additives. Bakers' yeast catalyzes the reduction of ethyl 4-chloro-3-oxobutanoate to the corresponding @-hydroxyester. The stereoselectivity of the reduction, however, is low58s because the D-and L-enzymes of bakers' yeast display nearly equal activities toward the substrate. It is conceivable that the change in stereoselectivity that occurs on introduction of an additive stems from an interaction between the additive and one or more of the oxidoreductases. To gain information on how additives affect the stereoselectivity of the reduction, the activity of each enzyme was measured in the presence of varying amounts of such additives as methyl vinyl ketone and ethyl chloroacetate. The reaction is inhibited by such compounds. The inhibition patterns were determined from the Dixon plots (plots of l / v vs [I])shown in Figures 7 and 8. The results show that the mechanism of inhibition by methyl vinyl ketone is noncompetitive for all the (11) Davis, B. J. Ann. N.Y. Acad. Sei. 1964, 121, 404.
4782 J. Org. Chem., Vol. 56, No. 15, 1991
Nakamura et al.
60 1
1
*r7
60
L-Enrymr-2
60
-200
.lo0
100
0
200
-60
-40
20
0
-20
40
-5
-10
0
5
10
[I1 (mM)
Figure 7. Dixon plota for the inhibition by methyl vinyl ketone. 80 -,
80
D-Enzyme-2
f
60-
60
3
40-
40
-
20
s 'e-
20
0 -20
01
:IO0
-50
0
50
100I
01
.10
0
10
20
")
Figure 8. Dixon plots for the inhibition by ethyl chloroacetate. Table 111. Inhibition Constants and Inhibition Patterns of the Enzymes inhibition inhibitor enzyme Kp (mM) pattern methyl vinyl ketone enzyme-2 154 noncompetitive noncompetitive L-enzyme-1 50.6 ~-enzyme-2 6.38 irreversibleb ethyl chloroacetate
~-enzyme-2 89.2 noncompetitive L-enzyme-1 very large ~-enzyme-2 4.42 mixed
OErrors in Ki are estimated to be lese than &lo%. *Irreversibility was confirmed only for this enzyme. Partial irreversibility was observed with the other enzymes. However, no definite conclusions could be drawn.
enzymes and that by ethyl chloroacetate is also noncompetitive for ~-enzyme-2but is of a mixed type for Lenzyme-2. The activity of L-enzyme-1 is little affected, even in solutions saturated with ethyl chloroacetate (ca. 150 mM). The kinetic constants obtained from the Dixon plota and the inhibition patterns are shown in Table 111. Discussion There are several oxidoreductases present in bakers' yeast. For example, Furuichi et al.12 isolated a &hydroxy ester oxidoreductase therefrom. The purified enzyme catalyzes the reduction of benzyl 2-methyl-3-oxobutanoatete to the corresponding L-&hydroxy ester, and thus is an L-enzyme. There is no doubt, however, that this enzyme differs from the L-enzymes described here, because the latter catalyze the reduction of ethyl 2-methyl-3-oxobutanoate,13 whereas the former does not. Another oxidoreductase was isolated by Furuichi et al." from Saccharomyces fermentati. This enzyme also differs from the L-enzymes described here, both in molecular weight and substrate specifi~ity.'~Shieh et al.16 isolated three dominant competing 8-hydroxy ester oxidoreductases from bakers' yeast. All three catalyze the reduction of ethyl (12) Furuichi, A.; Akita, H.; Mataukura, H.; Oiehi, T.; Horikoehi, K. Agric. Biol. Chem. 1986, 49, 2563-2670.
(13) Nakamura, K.; Kawai, Y.; Miyai, T.; Hoda, H.; Nakajima, N.; Ohno, A. Bull. Chem. SOC.Jpn., in press.
4chloro-3-oxobutanoate. Studiesle of the stereospecificity of the reductions catalyzed by these enzymes revealed that two, one of which was attributable to fatty acid synthetase (FAS), give the D-hydroxy ester, and thus are D-enzymes, whereas the thrid gives the L-hydroxy ester, and thus is a~ L-enzyme. The D-enzyme-2 described here may be identical with the FAS of Shieh et al. because the stereospecificities, molecular weights, and kinetic constanta of the two are very similar. The other D-enzyme described by Shieh et al. also has a molecular weight and kinetic constants similar to the Denzymel described here. Thus, the latter two enzymes may be the one and the same. However, the tenzyme of Shieh et al. is distinct from the L-enzymes described here. Why this is so is not clear. It may reflect the fact that the samples of bakers' yeast used in the two studies were obtained from different sources. Because the presence of a,@-unsaturatedcarbonyl compounds leads to an increase in the yield of the D product and also retards the rate of reduction? there is no doubt that such compounds inhibit the activities of the Lenzymes. The resulta of rate studies, shown in Figure 7 and listed in Table 111, clearly support this conclusion. An expression for the relative rate of reduction (a) in the presence of a noncompetitive inhibitor is'' a = Ki/Wi + [I]) Under preparative conditions, where the concentration of methyl vinyl ketone is ca. 50 mM, the relative rates of reduction catalyzed by ~-enzyme-2,L-enzyme-1, and Lenzyme-2 are 0.75,0.50, and 0.11, respectively. In other words, when the sum of the activities of the D- and Lenzymes is taken into consideration, it can be concluded that, in the presence of methyl vinyl ketone, the Denzyme retains 76% of ita original activity, whereas the L-enzymes (14)
Furuichi, A.; Akita, H.; Mataukura, H.; Oishi, T.; Aorikwhi, K.
Agric. Biol. Chem. 1987, 51, 293-299. (15) The reduction of ethyl 2-methyl-3-oxobutanoclteis not c a t a l y d by the enzyme isolated from S. fermentati. (16) Shieh, W.-R.; Gopalan, A. S.; Sih, C. J. J. Am. Chem. Soc. 1986, 107. 299.1-2MA. (17) Segel, I. H. Enzyme Kinetics; Wiley: New York, 1976; pp 125-136.
J. Org. Chem. 1991,56,4783-4791 Scheme I1
.yz;:
D
;E%%:;
L
retain only 14%. Thus, ~-enzyme-2is virtually solely responsible for the catalysis of the reduction. Consequently, the D product is afforded stereoselectively. The introduction of an alkyl a-haloacetate shifts the stereoselectivity of the reduction toward formation of the L product? It is believed, therefore, that such compounds inhibit the activity of D-enzyme. However, as the results shown in Figure 8 and listed in Table I11 demonstrate, ethyl chloroacetate inhibits not only the activity of Denzyme-2 but also that of ~-enzyme-2.On the other hand, the activity of L-enzyme-1is unaffected by this additive. So, it can be concluded that, in the presence of ethyl chloroacetate, the reduction by bakers' yeast is catalyzed only by L-enzyme-1. Of course, the results obtained from in vitro studies do not satisfactorily account quantitatively for the results obtained in vivo. Because the location of the enzymes
4783
within the yeast cell, the permeability of the cell wall toward organic reagents, and many other complex biological phenomena must be taken into consideration to understand the behavior of living microbes, the difference in the results of in vitro and in vivo studies is not surprising. Nevertheless, at least qualitatively, it is apparent that the activities of the enzymes of bakers' yeast are specifically inhibited by certain compounds. Thus, the stereospecificity of reductions by bakers' yeast can be changed. The effects of additives on the stereoselectivity of the reduction can be summarized by the reactions shown in Scheme 11. Studies of the enzymatic reduction of other substrates and the application of the purified enzymes described here to organic synthesis are now in progress. The results will be reported elsewhere.
Acknowledgment. We thank the Ministry of Education, Science and Culture for financial support (Grantsin-Aid No. 01470022,01303007, and 02303004). Registry No. wp-keto ester reductase, 114705-02-1;L-a-keto ester reductase, 114705-03-2;ethyl 4-chloro-3-oxobutanoate, 7894-4; ethyl chloroacetate, 105-39-5.
Calixarenes. 25. Conformations and Structures of the Products of Arylmethylation of Calix[l]arenes C. David Gutsches and P. Amruta Reddy Department of Chemistry, Texas Christian University, Fort Worth, Texas 76129 Received June 25,1990
The arylmethylations of calix[4]arenes reported in this paper serve as companion pieces to an earlier study of the effects of reaction conditions and the structure of the derivatizing agent on the conformational outcome of the aroylation of calix[4]arenes. In contrast to the aroylation reaction, where a significant and predictable relation is observed between the para substituent of the aroyl chloride and the ratio of conformers formed, the arylmethylation reaction shows only a small and much lees easily predictable dependence of the conformer ratio on the para substituent of the arylmethyl halide. Also, whereas the products of aroylation are the cone and/or 1,3-altemateconfprmers, those of arylmethylation are the cone and/or partial cone conformers. While no rationale has yet emerged to explain this difference, a study of the benzylation of dibenzyl and tribenzyl ethers of 5,11,17,23-tetra-tert-butyl-25,26,27,28-tetrahydroxycalix[4]arene has established that the conformations of the tetraethers are not completely established until the third step in some cases and the fourth step in others.
Calixarenes are conformationally mobile macrocyclic compounds that are the focus of considerable attention because of their ability, when appropriately functionalized, to serve as ionophores and enzyme mimics.' The possibility for conformational isomerism in the calix[l]arenes was first recognized by Megson2and Ott and Zinke;3the concept waa sharpened by Comforth and co-workers;' and the process of conformational interconversion has now been studied in detail by several groups of researchers including K h m e r e r et al.P*6Munch,' Gutsche et al.,Bpg ~~~~~
~
~
~
(1) Gutache, C. D. Calixarenes; Stoddart, J. F., Ed.;Royal Society of Chrmirtry: London, 1989. (2) M r p n , N. R. L. Oerterr. Chem.-Ztg. 1968,64,317. (3) Ott, R.; Zinkr, A. Oerterr. Chem.-Ztg. 1964,66, 166. (4) Comforth, J. W.; DArcyHart, P.; NicholL, G. A,; Reee, R. J. W. Br. J . Pharmacol. 1966, 10, 73. (6) KAmmerer, H.; Happel, 0.;C-r, F. Makromol. Chem. 1972,162, 179. (6) Happel, G.; Mathiasch, B.; Khmerer, H. Makromol. Chem. 1976, -176 .-, R--R I .I . (7) Munch, J. H. Makromol. Chem. 1977,69,178. (8) Gubche, C. D.; Bauer, L. J. Tetrahedron Lett. 1981, 22, 4763. (9) Bauer, L.J.; Gubchr, C. D. J . Am. Chem. SOC.1986, 107, 6052. 6063.
-
Shinkai et al,,l*l3 Ungaro et al.,14 and Reinhoudt et al.16 The four %p-downn conformers that are possible have been designated as cone, partial cone, 1,2-alternate, and 1,3-alternate, as depicted in Figure 1. Upon replacement of the phenolic hydrogens with sufficiently large groups, the calix[rl]arenes become conformationally inflexible, existing as discrete entities in one or another of the conformations.16J7 A study of the aroylation of calix[l]arenes carried out in this laboratory'* (10) Araki, K.; Iwamoto,K.; Shinkai, S.;Matauda, T. Chem. Left. 1989, 1747. (11) Araki, K.; Shinkai, 8.;Matsuda, T. Chem. Lett. 1989, 581. (12) Shinkai, 5.; Iwamoto,K.; Araki, K.; Matauda, T. Chem. Lett.
----. (13) Iwamoto,K.; Fujimoto, K.; Matauda, T.;Shinkai,S. Tetrahedron
1940.1263. ~ - - -
Lett. 1990,49, 7169. (14) Alfieri, C.; Dradi, E.; Pochini, A.; Ungaro, R. Gazz. Chim. Ital. 1989. - - - -, 119.336. - - -,- - -. (15) Gmtenhub, P. D. J.; Kollman, P. A.; Groenen, L. C.; Rehhoudt, D. N.; van Hummel, G. J.; Ugozzoli, F.; Andrwtti, G. D. J. Am. Chem. SOC.1990, 112,4165. (16) Gutache, C. D.; Dhawan,B.; No, K. H.; Muthukrbhnan, R. J. Am. Chem. SOC.1981,103,3782. (17) Bocchi, V.; Foina, D.; Pochini, A,; Ungaro, R.; Andreetti, 0. D. Tetrahedron 1982,38,373.
0022-3263/91/1956-4783$02.50/0(0 1991 American.Chemica1 Society
